mapclassify.HeadTailBreaks

class mapclassify.HeadTailBreaks(y)

Head/tail Breaks Map Classification for Heavy-tailed Distributions.

Parameters:

ynumpy.array

\((n,1)\), values to classify.

Notes

Head/tail Breaks is a relatively new classification method developed for data with a heavy-tailed distribution.

Implementation based on contributions by Alessandra Sozzi <alessandra.sozzi@gmail.com>.

For theoretical details see [Jia13].

Examples

>>> import mapclassify
>>> import numpy
>>> numpy.random.seed(10)
>>> cal = mapclassify.load_example()
>>> htb = mapclassify.HeadTailBreaks(cal)
>>> htb.k
3

>>> htb.counts.tolist()
[50, 7, 1]

>>> htb.bins
array([ 125.92810345,  811.26      , 4111.45      ])

>>> numpy.random.seed(123456)
>>> x = numpy.random.lognormal(3, 1, 1000)
>>> htb = mapclassify.HeadTailBreaks(x)
>>> htb.bins
array([ 32.26204423,  72.50205622, 128.07150107, 190.2899093 ,
       264.82847377, 457.88157946, 576.76046949])

>>> htb.counts.tolist()
[695, 209, 62, 22, 10, 1, 1]

Attributes:

ybnumpy.array

\((n,1)\), bin IDs for observations.

binsnumpy.array

\((k,1)\), the upper bounds of each class.

kint

The number of classes.

countsnumpy.array

\((k,1)\), the number of observations falling in each class.

__init__(y)

Methods

__init__(y)
find_bin(x)	Sort input or inputs according to the current bin estimate.
get_adcm()	Absolute deviation around class median (ADCM).
get_fmt()
get_gadf()	Goodness of absolute deviation of fit.
get_legend_classes([fmt])	Format the strings for the classes on the legend.
get_tss()	Returns sum of squares over all class means.
make(*args, **kwargs)	Configure and create a classifier that will consume data and produce classifications, given the configuration options specified by this function.
plot(gdf[, border_color, border_width, ...])	Plot a mapclassifier object.
set_fmt(fmt)
table()
update([y, inplace])	Add data or change classification parameters.

Attributes

fmt